Two Faces of Collaboration: a Critical Perspective on Effects of Collaboration in Learners'
Two Faces of Collaboration: A Critical Perspective on Effects of Collaboration in Learners' Corpus Consultation
ABSTRACT
Collaborative learning has been increasingly incorporated into learners' corpus consultation as a type of pedagogic mediation. However, studies found that collaborative learning has negative aspects as it may cause conflicts and power inequality among members. Given the two sides of collaboration, this study provides a critical investigation on the functions of collaboration in learners' corpus consultation. Through two collaborative corpus analysis sessions, three students of different L2 proficiency levels assisted each other’s understanding of corpus data to complete group activities. The students’ verbal and nonverbal behaviors during corpus analysis became the main source of investigation, complemented by the results of pre-/post-interviews and the researcher's direct observations. The findings of this study suggest that collaboration creates significant power inequality among the participants by gradually marginalizing the less capable student in the group from the learning process and increasing the psychological burden for the most capable student. In addition, collaborative work during corpus consultation could impede critical learning opportunities for individual students to pursue personal queries during collaboration. Despite some limitations, the findings of this study provide a realistic picture of collaboration in learners’ corpus consultation and suggest pedagogical implications to corpus-based instruction.
Keywords: learners' corpus consultation; collaborative learning; two-sides of collaboration; pedagogic mediation; critical perspective
INTRODUCTION
Learners' corpus consultation has brought many advantages in L2 learning such as enhanced linguistic awareness and learning autonomy (e.g., Boulton, 2009; Yoon, 2008), but the advantages come with serious difficulties for the students (e.g., Pérez-Paredes, Sánchez-Tornel, Alcaraz Calero & Jiménez, 2011). As Conrad (2005) indicated, students experienced difficulties in analyzing corpora because they needed "a number of technological and research skills: to use the computer, to handle what may at first appear to be an overwhelming amount of data, to make observations, to generalize from observations, and to assess the limitations of the generalization they derive" (Conrad, 2005, p. 402). Without appropriate guidance, students had to exert great time and energy in the data analysis process and were at high risk of analyzing the corpora incorrectly. Thus, researchers have noted that learners' corpus consultation should include a proper type of pedagogic mediation to make corpus data pedagogically appropriate (e.g., Widdowson, 2000).
Among several types of pedagogic mediation such as providing pre-selected and pre-edited data (e.g., Tribble & Jones, 1997), using computer-guided instruction (e.g., Curado Fuentes, 2003), and taking a guided inductive approach (e.g., Flowerdew, 2012), collaborative corpus consultation appear to have received a lot of attention with the increasing popularity of collaborative learning in L2 pedagogy (e.g., Flowerdew, 2008; Gavioli & Aston, 2001; O'Sullivan, 2007). In examining the pedagogic value of corpus data
ISSN: 1675-8021
that provoked a pragmatic reaction between students, Gavioli and Aston (2001) suggested the use of the corpus encouraged spoken interaction between students because they discussed their findings with each other in this situation. As students often reached different conclusions from the same concordances, the research pointed out that students could arrive at interpretations that were more comprehensive, or more generalizable, by comparing their analyses through collaborative group work. With the example of a group examining concordances of food and cibo (an Italian word for food), the authors showed that a group of students jointly hypothesized a difference in the meanings of the two words after discussion of their findings.
Similarly, O’Sullivan (2007) suggested that the learners' corpus consultation enhanced the language learning process, which was cooperatively constructed in a group. Advocating process-based as opposed to product-based learning, the author noted that corpus consultation agreed with the theoretical assumptions of process-oriented teaching and constructivist principles in L2 learning in that the corpus data offered the resources for students to co-construct their knowledge as a community.
Drawing on Vygotskian sociocultural theories, Flowerdew (2008) adopted peer response activities to use the corpus data for creating a contextual writing environment. The study showed that group activities in consideration of participants’ different language proficiencies and analytic abilities enabled co-construction of the learning process of corpus consultation, in that group members assisted each other’s development. In the scaffolding-type of activity, the author intentionally designed groups including students with different levels of proficiency. During the activity, the weaker students were able to build knowledge as well as independence in learning, as more proficient students shared their interpretations and understanding on the corpus data. The author reported success of group discussion activities in the learners' corpus consultation as a form of pedagogic mediation, which improved awareness on differences in registers. The study demonstrated the benefits of group work in corpus consultation to encourage the weaker students' productive dialogue.
However, collaborative group work in corpus-based instruction has not always generated positive results. The study of Vannestål and Lindquist (2007) showed an example of group activity in learners' corpus consultation. Incorporating corpus consultation into L2 grammar instruction, the researchers paired students and asked them to deduce grammar rules from corpus consultation. Then, each pair of students in a group of four (i.e., two pairs in one group) exchanged their findings on grammatical rules. However, it was interesting to find that the students felt significant degree of difficulties especially in pursuing each query and interpreting the concordances. In fact, a student even casted doubt on the benefits of corpus consultation in grammar learning. In conclusion, the authors acknowledge the need for a large amount of introduction and support in learners' corpus analysis, hinting that the peer collaboration might not be an effective pedagogic mediation for the learners' corpus consultation.
Given the findings of prior studies, it seems noteworthy that the benefits of collaboration may have a flip side to examine, especially from a psychological and motivational perspective. In particular, it is apparent that the benefits of collaboration in learners' corpus consultation have been celebrated in terms of how generalizable or successful the results of collaboration were (e.g., Flowerdew, 2008; Gavioli & Aston, 2001; O’Sullivan, 2007), but disregard the effectiveness of the collaboration in alleviating the learners' difficulties in corpus consultation. Thus, the mixed results on the efficacy of collaboration in learners' corpus consultation necessitated a qualitative investigation on the collaborative process of corpus consultation with focused attention on how collaboration actually functions in the process of learners' corpus consultation. In this sense, this study aimed to closely examine the process of collaborative corpus consultation to reveal the true
ISSN: 1675-8021
3functions of collaboration and to provide practical implications in learners' corpus consultation in L2 instruction.
TWO SIDES OF COLLABORATION IN LANGUAGE LEARNING
Collaborative learning involves the "mutual engagement of participants in a coordinated effort to solve the problem together" (Roschelle & Teasley, 1995, p. 70). Johnson and Johnson (1999) found that collaboration enabled students to develop communication skills as well as social relationships and group cohesion. In addition, students were able to engage in deeper level learning, develop critical thinking skills, and retain instructed information in the long term through collaborative learning (e.g., Garrison, Anderson & Archer, 2001). Also, the collaboration facilitated learner's co-construction of knowledge and development of overall competence of language (e.g., Baleghizadeh & Arab, 2011; Keen, 1992). When engaging in a web search task, in particular, peer collaboration encouraged students to express their ideas, guided the search process, and regulated search results (Lazonder, 2005).
Despite the benefits of collaborative learning, successful collaboration cannot be guaranteed at all times. Collaboration includes negotiation and competition, the process of which creates conflicts and power inequality among the participants (e.g., Chan & Chen, 2010). In particular, inequality among group members seems to be a natural consequence of competition between members with different levels of ability in collaborative learning. For instance, in an examination of two non-native English speaking students' (NNS) experiences in mixed groups of native English speaking students (NS) and NNS in English university courses, Leki (2001) found that NNS were kept aware of their lack of power and control in collaborative group work due to their limited linguistic ability. Similarly, Cheng (2013) reported that the co-writing process of a mixed group of NS and NNS created a noticeable power inequality between participants, significantly caused by the lack of the NNS’s linguistic ability. The findings of the studies suggested a strong possibility that the less capable members within a group of NNS may be as vulnerable as the NNS in the mixed group of NS and NNS. In this sense, collaborative learning may create an environment that discourages and demotivates the less capable members to make contributions to group work.
In fact, studies on affective and motivational aspects of collaboration have reported pitfalls of collaboration in the learning process (e.g., Järvenoja & Järvelä, 2009). The process of collaboration may include conflicts due to students' different characteristics, goals, and demands, which cause negative emotions and discouragements (e.g., Järvelä, Lehtinen, & Salonen, 2000). In particular, Järvenoja and Järvelä (2009) noted that the process of co-construction of knowledge may entail strong motivational challenges, discouraging the learners to participate in collaboration. In this sense, collaboration in learning may demotivate some learners to contribute to the group task, causing psychological distress in relationships between members and disturbing the learning process itself.
Based on the findings of previous research on positive and negative sides of collaboration, the present study presents findings based on a critical observation on functions of collaboration in the process of learners' corpus consultation. This study analyzes three sources of data, including verbal and nonverbal communications during group work, findings from pre- and post-interviews, and the instructor's observation notes, to examine the process of learners' collaboration from multiple perspectives. Through two tasks of collaborative corpus consultation by three students at different L2 proficiency levels, this study identifies critical functions of collaboration, concerning the instructional, motivational, and psychological aspects of learning. The description of the collaborative corpus consultation provides a realistic picture of collaboration in learners' corpus analysis, which would deepen our knowledge on the collaborative process in corpus analysis.
ISSN: 1675-8021
4METHODOLOGY
PARTICIPANTS
Three female students were recruited through an online community website of a university in Seoul, Korea. On this website, many students exchanged opinions about their school life and gained information about part-time jobs. Finding a posting to recruit experiment participants, the students gave contact to the researcher with interests in learning English. They varied in terms of age, major, linguistic proficiency level based on TOEIC (Test of English for International Communication) score, and experience studying abroad (Table 1). In terms of linguistic proficiency level, this study utilized categories from A to E based on the TOEIC score by ETSi. S1 is in level B (TOEIC score higher than 730 and lower than 860), and S2 and S3 are in level A (860 or higher). It is expected that the students different linguistic proficiency levels would allow natural flow of knowledge between group members during the corpus consultation work.
TABLE 1. Description of participants
Students / S1 / S2 / S3Age / 30 / 23 / 21
Major / International Relations / Business Administration / Japanese
TOEIC score / 775 / 885 / 910
Linguistic proficiency level / B / A / A
Self-reported computer skills / medium / medium / medium
Six months in
Experience abroad / None / None / Canada as an
exchange student
Experience in corpus linguistics / None / None / None
Among the three participants, S1 had the lowest linguistic proficiency level based on her TOEIC score. During the pre-interview, she noted that despite her major (International Relations), which required a good command of English, her English is not good enough to understand English lectures. Thus, she felt a strong need to study English and keep applying for TOEIC tests to improve her score. In addition, S1 was the oldest member as she had entered this university after graduating from a different college. S2 majored in Business Administration and had strong interests in improving her English because her major required a high English proficiency. During the introductory session of this study, she seemed thrilled as she thought the corpus consultation is an innovative way of learning English. S3 was the youngest student, but had the highest TOEIC score. She had spent six months in Canada, while the other two had no experience of studying abroad. S3 majored in Japanese, but had high English proficiency. During the corpus consultation, she read and comprehended concordance lines the fastest, and was the most accurate among the three participants.
DATA COLLECTION PROCEDURE
The introductory class started with a written survey asking for the participants' personal information. The survey also asked students to write sets of synonymous expressions whose usage and semantic differences seemed to be indiscernible to them such as affect/effect/influence and hope/wish. The sets of synonymous expressions were used in designing two tasks of corpus consultation in this study. After the survey, the students took an hour lecture about the basic concepts of corpus linguistics and how to use the Lextutor, a web-based system for exploring online corpora. After the lecture, the students were given
ISSN: 1675-8021
5hands-on activities to use the Lextutor and analyzed search results. At the end of the class, students were interviewed to share their opinions and expectations on using corpus consultation in learning English.
The following day, the students worked as a group to complete a task by using collaborative corpus analysis. The students were given one computer to share, which was to encourage verbal and nonverbal communication among members. Before the start of corpus analysis, each student was given a worksheet to fill in, which asked the students to discern semantic and functional differences of two sets of synonymous expressions (affect/effect/ influence and travel/journey/trip). Through group discussion on concordance lines of search terms on Lextutor, the students created hypotheses on differences among the expressions, similar to the students who did this in the study by Gavioli and Aston (2001).
The third day’s session required students to analyze corpus data as a part of an essay writing activity (i.e., translation as a standardized task of writing in this study). The reason for including the writing activity in the experiment was that during the pre-interview, the students were expected to use corpus data as an L2 writing reference. The task consisted of two parts: the first was to consult corpus data to discern semantic and functional differences of three sets of synonymous expressions (even as/even though/during, worth/worthy, hope/wish): and the second was to translate a short Korean passage into English by using appropriate words among the given synonymous expressions. For successful completion of the task, it is important for students to identify differences among the synonymous expressions and choose appropriate ones for translation. The answers that students provided in the worksheet could vary since each student was given individual worksheet, but the process of corpus analysis was collaborative as they were encouraged to discuss their ideas on the search results.